Advances in computing and machine learning (ML) methods have led to a rapid rise in artificial intelligence (AI) research and applications in many fields. AI research benefitted from advances in computation hardware, collection and distribution of large data sets, and proliferation of software techniques. AI techniques include ML for provable results, deep learning for data exploration, reinforcement learning for control, and active learning for adaptive systems. Likewise, AI algorithms can handle large amounts of data, construct unknown representations, and provide a direct link between data and classification for decision making. These unmatched capabilities have been seen as a path to solving hard engineering problems, including that of structural health monitoring (SHM). SHM consists of automating the condition assessment task of civil, health, mechanical, and aerospace systems using measurements obtained from temporary or permanently installed sensors. Often, the systems of interest are geometrically large and/or technically complex, which complicates the development and application of physics-based methods. It follows that AI is seen as a key potential contributor enabling SHM in field applications for data-driven analysis. As with many research endeavors, many concepts using AI for SHM have been explored in the literature. Nevertheless, very few AI methods have been deployed in the context of SHM, which may be due to the lack of available data supporting their capabilities, limited integrated AI-SHM systems capable of providing results to users and operators with decision-making capabilities, or certification of AI methods for safety-critical applications. The objective of this Roadmap publication is to discuss the integration of AI at the system level enabling SHM, including associated challenges and opportunities such as those found in common metrics of concern (e.g. transparency, interpretability, explainability, security, certifiability, etc), with a particular focus on providing a path to research and development efforts that could yield impactful field applications. The overview of available methods and directions will provide the readers with applicability of AI for certain SHM designs (software), availability of common data sets for further AI comparisons (data), and lessons learned in implementation (hardware).

Autonomous underwater vehicles with greater maneuverability, efficiency, and resiliency are needed to meet the challenges of exploring and monitoring the underwater world, so we look to underwater creatures to uncover what makes them such excellent swimmers. Bio-inspired, soft robots can combine the performance of biological swimming with the robustness of soft construction, where the ideal robot has a jointless, flexible body with embedded muscles just like real fish. In this paper, we present a continuously deformable robotic trout with embedded electro-hydraulic hydraulically amplified, self-healing electrostatic artificial muscles, experimentally characterize its swimming kinematics, and report a reduced order numerical model which predicts the robot fish’s natural frequencies and mode shapes. We characterized the robot’s 3D full body swimming kinematics while submerged in water with digital image correlation. The soft robot undergoes whole body bending in response to internal muscle actuation and yields kinematics comparable to biological trout. Tail beat velocity was measured at the first three observed natural frequencies with a maximum of 69 mm s⁻¹ corresponding to a caudal fin trailing edge displacement of ±10 mm. We derive a beam-based fluid structure interaction (FSI) model which predicts swimming kinematics in response to embedded muscle forces and includes the effects of nonlinear vortex and convective forces on the robot’s body. The nonlinear FSI model predicted the first three damped natural frequencies within 5% error and mode shapes which correlated with the experimental data. This paper contributes the design, fabrication, and characterization of a solid-state robotic trout featuring whole-body flexibility and embedded actuation through numerical modeling and experimental analysis.

Structure-Borne Traveling Waves (SBTWs) have significant promise as a means of underwater propulsion, solid-state motion, and drag reduction. However, there is limited understanding over how to tailor SBTWs for these applications. This study establishes guidelines and a procedure for tailoring SBTWs on thin-walled two-dimensional (2D) surfaces. These guidelines were specifically crafted to yield SBTWs favorable for propulsion and solid-state motion: propagation along a uniform direction and with consistent amplitude. SBTWs are steady-state traveling waves generated by taking advantage of a surface’s modal properties (i.e. mode shapes). This novel mechanism provides significant benefits including a minimal actuation footprint, a wide frequency bandwidth, and active variability (propagation direction, wavelength, etc). However, this mechanism also introduces a dependence on the mode shapes and actuator location. Thus, to tailor SBTWs favorable for propulsion the relationship between SBTWs, the mode shapes, and actuator locations must be understood. Eight Guidelines governing this relationship are established. These Guidelines are demonstrated using SBTWs generated on a previously validated model of a 2D plate. Among others, it is shown that the frequency bandwidth can be broken into distinct frequency regions, and each region can be classified as yielding favorable or unfavorable SBTWs. A multi-step procedure is then introduced that enables SBTWs to be tailored on any thin-walled surface. Numerically or experimentally applicable, the actuator configuration can be defined to tailor SBTWs with a specific propagation direction; all that is required is knowledge of the mode shapes. Finally, a case study is presented in which the guidelines and procedure are implemented to tailor SBTWs on a trapezoidal surface. Two different sets of SBTWs are tailored, each propagating in different directions and with various wavelengths. Thus, the established guidelines and procedure enable SBTWs to be tailored for specific applications, such as propulsion and solid-state motion.

This study identifies hydraulically amplified self-healing electrostatic (HASEL) transducers as electricity generators, contrary to their conventional role as actuators. HASELs are soft, variable-capacitance transducers inspired by biological muscles which were developed to mimic the flexibility and functionality of natural muscle tissues. This research characterizes HASELs as generators by reversing their energy conversion mechanism—generating electricity through mechanical deformation. The study assesses the practical laboratory performance of HASELs by analytic modeling and experimental evaluation. Outcomes of the study include the following: (i) up to 2.5 mJ per cycle per 50 mm wide HASEL pouch of positive net energy generation in experimental testing—corresponding to an energy density of 2.0 mJ cm⁻³; (ii) a maximum theoretical energy density of 4.2 mJ cm⁻³; (iii) the electromechanical characteristics governing efficient conversion; and (iv) design considerations to enhance HASEL generator performance in future applications. This study broadens HASEL’s applicability and utility as a multi-functional transducer for renewable energy and general adaptive electricity generation.